hepatitisbandhepatitiscinfectionbiomarkersand...

Hindawi Publishing CorporationHepatitis Research and TreatmentVolume 2011, Article ID 582945, 10 pagesdoi:10.1155/2011/582945

Research Article

Hepatitis B and Hepatitis C Infection Biomarkers andTP53 Mutations in Hepatocellular Carcinomas from Colombia

Maria-Cristina Navas,1 Iris Suarez,1 Andrea Carreno,1 Diego Uribe,1

Wilson Alfredo Rios,1 Fabian Cortes-Mancera,1 Ghyslaine Martel,2 Beatriz Vieco,3

Diana Lozano,3 Carlos Jimenez,3 Doriane Gouas,2 German Osorio,1, 3 Sergio Hoyos,1, 4

Juan Carlos Restrepo,1, 4 Gonzalo Correa,1, 4 Sergio Jaramillo,4 Rocio Lopez,5

Luis Eduardo Bravo,6 Maria Patricia Arbelaez,7 Jean-Yves Scoazec,8

Behnoush Abedi-Ardekani,2 Regina M. Santella,9 Isabelle Chemin,10 and Pierre Hainaut2

1 Grupo de Gastrohepatologia, Facultad de Medicina, Universidad de Antioquia, Medellın, Colombia2 International Agency for Research on Cancer, 150 Cours Albert Thomas, 69372 Lyon, France3 Departamento de Patologıa, Facultad de Medicina, Universidad de Antioquia, Medellın, Colombia4 Hospital Pablo Tobon Uribe, Medellın, Colombia5 Departamento de Patologia, Fundacion Santa Fe de Bogota, Bogota D.C., Colombia6 Departamento de Patologıa, Facultad de Salud, Universidad del Valle, Cali, Colombia7 Grupo de Epidemiologia, Facultad Nacional de Salud Publica, Universidad de Antioquia, Medellın, Colombia8 Service d’Anatomie Pathologique, Hopital Edouard Herriot, 69437 Lyon, France9 Mailman School of Public Health, Columbia University, NY 10032, USA10INSERM U871, 69424 Lyon, France

Correspondence should be addressed to Maria-Cristina Navas, [email protected]

Received 23 March 2011; Accepted 22 August 2011

Academic Editor: Patrick Soussan

Copyright © 2011 Maria-Cristina Navas et al. This is an open access article distributed under the Creative Commons AttributionLicense, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properlycited.

Hepatocellular Carcinoma (HCC) is a leading cause of cancer-related death worldwide. Globally, the most important HCC riskfactors are Hepatitis B Virus (HBV) and/or Hepatitis C Virus (HCV), chronic alcoholism, and dietary exposure to aflatoxins. Wehave described the epidemiological pattern of 202 HCC samples obtained from Colombian patients. Additionally we investigatedHBV/HCV infections and TP53 mutations in 49 of these HCC cases. HBV biomarkers were detected in 58.1% of the cases; HBVgenotypes F and D were characterized in three of the samples. The HCV biomarker was detected in 37% of the samples whileHBV/HCV coinfection was found in 19.2%. Among TP53 mutations, 10.5% occur at the common aflatoxin mutation hotspot,codon 249. No data regarding chronic alcoholism was available from the cases. In conclusion, in this first study of HCC andbiomarkers in a Colombian population, the main HCC risk factor was HBV infection.

1. Introduction

Primary liver cancer is the third leading cause of cancerdeath. Moreover, it is the fifth and eighth most frequentcancer among men and women worldwide, respectively.The most common histological type of liver cancer ishepatocellular carcinoma (HCC) accounting for 80 to 90%of the cases [1].

HCC incidence is highly variable among geographicregions depending on the prevalence of risk factors and theincidence of liver cirrhosis; actually, 70 to 90% of HCCcases develop from cirrhotic liver. Major risk factors of HCCinclude Hepatitis B Virus (HBV) and/or Hepatitis C Virus(HCV) infection and heavy alcohol consumption. In fact,chronic HBV and HCV infections have been recognized asliver carcinogens with an imputable fraction of at least 75%

2 Hepatitis Research and Treatment

of HCC cases; moreover, it has been estimated that HBVis responsible for 50 to 80%, whereas HCV is associated to10 to 25% of HCC cases. Other environmental and geneticHCC risk factors include dietary exposure to aflatoxins, dia-betes, obesity, nonalcoholic steatohepatitis, and hereditaryhemochromatosis [1–3].

The burden of HCC is growing in different continents.Central and South America were in the past known as low-incidence liver cancer regions. However, according to the lastpublished GLOBOCAN analysis, the incidence rates of livercancer in these countries correspond to low and intermediateincidence.

Colombia is a country of relatively low incidence ofliver cancer with incidences of primary liver and bile ductcancers of 3.1/100,000 in males and 2.7/100,000 in females.However, there is only one active cancer registry in thecountry, based in Cali city, an urban area; nevertheless,whether this situation is representative for the country as awhole is unknown [4]. Additionally, the national mortalityregistry reported around 1,300 deaths from malignant liverand intrahepatic bile ducts cancer that corresponds to amortality rate of 3.23 and 3.09/100 000 in men and women,respectively [5]. So far, there is no study assessing thegeographic variations in incidence or risk factor of chronicliver disease and liver cancer in Colombia.

Latin American data about HCC risk factors are limited.The first recent prospective study of HCC etiology in 9Latin American countries showed that the primary risk factorwas chronic HCV infection (30.8%), followed by chronicalcoholism (20.4%), and chronic HBV infection (10.8%) [6].Although HCV infection is the most important HCC riskfactor in Argentina, Mexico, and Brazil, regional differenceshave been described between northern and southern statesin Brazil. Indeed, HBV infection is the most prevalent riskfactor in northern states in Brazil, as in Peru [7–14].

According to the World Health Organization, Colombiahas a moderate endemicity for HBV; although there are sev-eral epidemiological patterns given the geographic, ethnic,cultural, and socioeconomic status of the population. Datafrom the Colombian National Institute of Health indicatethat, in 2007, a seroprevalence of HBsAg of 0.27% (range0.08–1.27) was found in 1573 blood bank samples fromacross the country. In some rural areas, such as Amazonasstate, rates of chronic HBV carriage over 5% have beenreported [15, 16].

Although the prevalence of HCV infection in the generalpopulation in Colombia is unknown, the WHO estimates aprevalence between 1 to 2.5% for this country, consideringthe data from the National Blood Banks Unit of theColombian National Institute of Health. Indeed, while theseroprevalence of HCV in blood donors was 0.7–1% in 1993–1996 and 0.5% in 2002, in a cohort of 500 multitransfusedpatients recruited from the two largest cities in Colombia,Bogota and Medellin, the HCV prevalence was 9% [17].

Data on exposure to aflatoxins, a class of mycotoxincontaminating traditional foodstuff in tropical countries, areeven scarcer [18]. A survey of aflatoxin contamination inselected Colombian foods was conducted over a 12-monthperiod on a total of 248 samples collected in supermarkets,

retail stores, and stock centres [19]. Aflatoxins were detectedin 22 samples, including 14 of 109 samples of corn and cornproducts and 4 of 40 samples of rice and rice products.Twelve of the 22 positive samples exceeded the maximumtolerable level of AFB1 adopted by most countries (5 ng/g),including 10 samples of corn and corn products. Given thatcorn is part of the common diet of Colombian inhabitants;it is likely that AFB1 may represent a significant exposureat least in a fraction of the population. Finally, the roleof chronic alcoholism (mostly in the form of cane sugaralcohol) may also be significant [20–23].

In the present study, we describe the sociodemographicvariables of 202 HCC cases, who attended four referenceinstitutions in Colombia during the period 2000–2007, and,for the first time, the prevalence of biomarkers in a seriesof 49 HCC cases. We report that the HBV biomarker wasdetectable in 58.1% of the cases and the HCV biomarker in37%. Among TP53 mutations, 10.5% occur at the commonaflatoxin mutation hotspot, codon 249, although G12457T(exon 5) and G13804A (exon 8) were present in 2.9% ofthe HCC samples. Unfortunately, data on chronic alcoholismwas not available from the cases. These results suggest thatthe principle HCC risk factor in this Colombian populationis HBV infection and low to moderate of AFB1 exposure.

2. Materials and Methods

2.1. Liver Samples. HCC samples were obtained fromarchived cases in the Departments of Pathology of fourinstitutions in the three largest cities in Colombia, Bogota,Medellin, and Cali, during the period from 2000 to 2007.The institutions correspond to Fundacion Santa Fe de Bogota(47 cases), Hospital Pablo Tobon Uribe (31 cases), Facultadde Medicina, Universidad de Antioquia (114 cases), andHospital Universitario del Valle (10 cases).

From 202 HCC cases registered at the archives, 49paraffin-fixed liver samples were available for immuno-chemistry and molecular biology assays. The histologicalpattern and grade of tumor differentiation (Edmonson andSteiner grading system) was assigned by two independentpathologists.

2.2. Immunochemistry Analysis

2.2.1. p53. The isoforms of p53 were detected on deparaf-finized tissue sections using standard protocols with CM1antibody (rabbit polyclonal immunoglobulin G antihu-man p53, 1/500, Novacastra Laboratories Ltd., Newcas-tle, UK). The antibodies were detected using biotiny-lated immunoglobulin G, streptavidine-peroxidase, anddiaminobenzidine-based detection (Vector Laboratories,Inc., Burlingame, Calif, USA).

2.2.2. HBx and Core HCV. After deparaffinization andrehydratation antigen retrieval was applied by vaporizerin Target Retrieval Solution pH 6.0 (Dako). As primaryantibody Core HCV monoclonal antibody (anti-Core HCV

Hepatitis Research and Treatment 3

aa70–90 CHEMICON, Millipore) or HBx monoclonal anti-body Hepatitis B Virus X-Protein (Trans-Activator X GeneProduct) monoclonal antibody (Clon 227, CHEMICON,International, Inc.) were used at a dilution 1 : 50. The kitultravision LP detection System HRP Polymer and DABPlus Chromogen (Lab Vision corporation) was used for thedetection of HBV and HCV antigens.

2.2.3. AFB1-DNA Adducts. The standardization of adductsdetection was performed using liver tissue sections fromrats treated with AFB1. The liver tissue sections werelabeled using the antibody highly specific for AFB1-Fapyadducts 6A10, developed and characterized by Hsieh etal. [24]. Briefly, the sections were first treated with 5 mMNa2CO3/30 mM NaHCO3 (pH 9.0) to open the guanineadducts ring. The antigen retrieval was performed usingcitrate buffer pH 6.0 (Dako). The liver sections were thentreated with RNase A (100 µL/mL, Fermentas, RNase A,DNase, and Protease-free), with Proteinase K for 10 min at37◦C (10 µL/mL, Gentra Puregene) and with NaOH 50 mMin 40% ethanol for DNA denaturation. Slides were thenincubated with the antibody 6A10 at a dilution 1 : 20 at 4◦Covernight. The reaction was detected using the kit ultravisionLP detection System HRP Polymer & DAB Plus Chromogen(Lab Vision Corporation).

2.2.4. DNA Extraction. DNA was extracted from 4 µmunstained paraffin sections. The sections were deparaffinedin xylene and ethanol. Then, the tumor tissue areas of interestwere scrapped into 1.5 mL sterile microcentrifuge. DNA wasextracted using QIAmp DNA Micro kit (Qiagen, Hilden,Germany), according to the manufacturer’s instructions.DNA extracts were stored at −20◦C.

2.2.5. TP53 Mutations. DNA was used for amplification ofexon 7 of TP53 gene with the primers flanking the exon(sense-333 ACTTGCCACAGGTCTCCCCAA and antisense-313 AGGGGTCAGCGGCAAGCAGA) as described else-where [25]. Briefly, the PCR was carried out in a volumeof 25 µL containing 5 µL of DNA, 1 U of Platinum TaqDNA polymerase High Fidelity (Invitrogen Carlsbad, USA),0.4 µM of each primer, dNTP (200 µM each), 1X HighFidelity Buffer Taq polymerase (Invitrogen), 0.5 mM ofMgSO4 (Invitrogen), and nucleases-free water (Amresco,Solon, USA). The PCR reaction involved a 15 min Hot-StarTaq polymerase activation at 95◦C followed by 45 cycles ofdenaturing at 94◦C for 30 sec, annealing at 60◦C for 30 secand extension at 72◦C for 30 sec, followed by a final extensionfor 10 min at 72◦C.

The specific G to T transversion at codon 249 of exon 7was analyzed by Restriction Fragment Length Polymorphism(RFLP). PCR products were digested by HaeIII restrictionendonuclease (Promega, Madison, USA). The fragmentswere visualized on 3% agarose gel stained with ethidiumbromide, eluted, reamplified, and sequenced by automatedsequencing (sequencer 3730xl). Additionally, all sampleswere analyzed by direct sequencing of PCR products cor-responding to TP53 exons 7 and also exons 5, 6, and 8 as

described elsewhere [26]. All results represent a minimum oftwo fully independent analyses.

2.2.6. Detection of HBV. HBx DNA sequences were ampli-fied by PCR. Briefly, the PCR was carried out in a20 µL volume containing Colorless GoTaq flexi buffer,MgCl2 1.5 mM, dNTP (200 µM each), primers DG-XF4(GGGACGTCCTTTGTCTACGT), and DG-X1R (GGGA-GACCGCGTAAAGAGAG) and 0.8 U of GoTaq DNA poly-merase (Promega). The PCR reaction involved a step at 95◦Cfor 2 min followed by 50 cycles of denaturing at 94◦C for45 sec, annealing at 62◦C for 45 sec and extension at 72◦C for45 sec, followed by a final extension for 7 min at 72◦C. PCRproducts were analyzed by electrophoresis on 2% Agarosegels and ethidium bromide staining.

2.2.7. HBV Genotyping. The small S gene fragment of HBVwas amplified using YS1-YS2 in the first round, and s3-s3as (319 nt) in a second round [27, 28], or hep3–hep33 asa unique round of PCR [29]. All sequences obtained werecompared with GenBank available sequences. Phylogeneticanalyses by Neighbour Joining and Maximum Likelihoodwere conducted with MEGA 5.1; this program was also usedfor tree representation.

3. Results

3.1. Characteristic of Study Population. During the period2000–2007, 192 HCC cases were diagnosed at FundacionSanta Fe de Bogota (23.3%), Hospital Pablo Tobon Uribe(15.3%) and Facultad de Medicina, Universidad de Antio-quia (56.4%). Additionally, 10 cases (5%) were diagnosedat Hospital Universitario del Valle during the period 2000–2004.

Among the total HCC cases, 36% were diagnosed infemales and 64% in males corresponding to a male/femaleratio of 1.8 : 1. The average age was 62 years, the median was61 years, and the age range was 22 to 90. The records ofHCC by age showed a higher frequency starting in the sixthdecade of life in both populations genders. No data fromthe cases were available on HCV or HBV status or alcoholconsumption.

The 49 HCC cases included in this study were classifiedaccording to the Edmonson-Steiner criteria as G1 well-differentiated (4.9%), G2 moderately differentiated (39%),or G3 poorly differentiated (56.1%). The trabecular type wasthe most frequent (56.5%) followed by solid (21.7%), mixte(13%), glandular (4.4%), and pseudoglandular (4.4%). Theclinicopathological characteristics of the HCC cases includedin this study are summarized in Table 1.

3.2. Mutations in TP53 Gene. The 249ser mutation was inves-tigated in HCC samples by RFLP followed by sequencing andby direct sequence of TP53 exon 7 (Figure 1). The presence ofa mutation was confirmed by both techniques in 4 (10.5%)of 38 HCC samples. The mutation was associated withoverexpression of p53 in two of these samples (10%–50% ofcells stained); for the other two cases, the immunochemistry


Table 1: Clinicopathological data of HCC cases diagnosed atfour health reference institutions during the period 2000–2007 inColombia.

Frequency (%) n = 49

Mean age 63.5 years

Age range 25–88 years

Gender female/male 43.2%/56.8%

Edmondson and Steiner’s grade

G1 4.9%

G2 39%

G3 56.1%

Histological type

Trabecular 56.5%

Solid 21.7%

Mixte 13%

Glandular 4.4%

Pseudo-glandular 4.4%

Biomarkers

HBV 58.1% (25/43)

HCV 37% (10/27)

Coinfection 19.2% (5/26)

AFB1-DNA adducts 0% (0/31)

TP53 gene

Exon 7 (249ser mutation) 10.5% (4/38)

Exon 5 (G12457T mutation) 2.9% (1/34)

Exon 8 (G13804A mutation) 2.9% (1/34)

analysis was not available. No additional mutations weredetected in any other position of TP53 exon 7.

The analysis by direct sequence of TP53 exons 5, 6, and 8revealed two point mutations in exon 5 (G12457T/V157F)and exon 8 (G13804A/C275Y) and one SNP in exon 6(A12708G/R213R) in three (8%) HCC samples of the 34without 249ser mutation. No accumulation of p53 proteinwas demonstrated in the HCC sample exhibiting the muta-tion G12457T. The assay of p53 protein by immunochem-istry was not available for the other two samples.

3.3. AFB1-DNA Adducts. A total of 31 liver tissue sampleswere analyzed for AFB1-DNA adducts by immunohisto-chemistry. This biomarker was detected in hepatocyte nucleiof treated rat liver tissue included in each of the assaysas a positive control; however, none of the HCC samplesanalyzed was positive for DNA adducts (Figures 2(a) and2(b)). Twenty-three HCC samples were analyzed for bothAFB1-DNA adducts and TP53 exon 7, but none of thesesamples was positive for the 249ser mutation.

3.4. HBV Infection Biomarkers. Considering that data onthe HBV infection status of the HCC cases analyzed wasunavailable, two biomarkers were included in order toidentify the cases associated with HBV infection. Twenty-five HCC samples (58.1%, 25/43) were positives for nuclearHBx protein by immunohistochemical detection (Figures2(c) and 2(d)) and/or HBx sequence detection by PCR

(nucleotides 1411–1549). From the 25 positives HCC cases,5 samples were positive for both biomarkers; eleven sampleswere positives for one of HBV biomarkers (HBx immunohis-tochemical detection or HBx PCR), while 9 samples positivewere analyzed for only one of the biomarkers.

The 249ser and G13804A (exon 8) mutations wereidentified in two samples positive for HBV biomarkers.Regrettably, the analysis of HBV biomarkers was not availablein one of the samples positive for the 249ser mutation.

3.5. HBV Genotypes. The small S gene fragment was analyzedby PCR and sequenced in 23 HCC samples positive for HBVbiomarkers (HBx protein and/or HBx PCR). However, itwas successfully sequenced in just three samples. One of thereasons for this limited number of samples characterized forthe viral genotype could be the quality of the DNA extractedfrom the paraffin-fixed liver tissues.

Phylogenetic study of the S sequence showed that twoisolates belonged to genotype F and one isolate to genotypeD according to the grouping with HBV prototype GenBanksequences. Similar topology was observed between treesgenerated by the different inference methods (data notshown). The identification of the subtype was not availablein these samples taking into account the limitations of thesize sequence (Figure 3).

3.6. HCV Infection Biomarker. The Core HCV protein wasdetected by immunohistochemistry in 37% of the samplesanalyzed (10/27); cytoplasmic staining was observed in allpositive HCC samples (Figures 2(e) and 2(f)). HBV/HCVcoinfection was demonstrated in 19.2% (5/26) of the liver tis-sue samples included in the assays; two were positive for bothbiomarkers of HBV and HCV infection, while the other threewere positive for HBx by immunohistochemical detection orHBx PCR, in addition to the Core HCV protein detection.

4. Discussion

This is the first study of HCC biomarkers carried out ina Colombian population. The health centers of this studyincluded two leading Departments of Pathology at thenational level (Fundacion Santa Fe de Bogota and Facultadde Medicina, Universidad de Antioquia) and two of the mostimportant hospitals in Medellin and Cali Cities (HospitalPablo Tobon Uribe and Hospital Universitario del Valle).

Although, the 202 cases recruited in these centersduring the period 2000–2007 do not represent the nationalregistries, the data obtained from these reference institutionscontributes to the knowledge of HCC epidemiology ofColombia.

The burden of HCC has demonstrated an increasingtrend over the past two decades in some regions of theworld [1]. Even in Latin America, previously consideredas a low incidence of liver cancer region, the data fromdifferent countries revealed increasing rates of HCC. Indeed,the mortality rates for primary liver cancer have increased inMexico from 4.1/100.000 inhabitants in 2000 to 4.7/100.000in 2006 [30]. A similar tendency is observed in Colombia; the


MW 1 2 3 4 5 6 7 8 9 10 11 12 13

(a)

A A C C G G A G T C C C A T C C T C

A A C C G G A G G C C C A T C C T C

Mut

WT

50

5040

(b)

Figure 1: TP53 249ser mutation in HCC cases from Colombia. (a) TP53 mutation at codon 249 was identified by restriction digestion.Presence of an undigested 158 bp fragment is indicative of mutation. Wild-type pattern: Lanes 2 (negative control: DNA from healthy donorlymphocytes), 4 to 8, 11, and 13 (HCC cases). Mutant pattern: Lanes 3 (positive control: DNA from PLC/PRF/5 cell line), 9, 10, and 12 (HCCcases). MW: molecular weight marker 100 bp. (b) Sequencing chromatograms of a mutant HCC case (a), showing the change from AGG toAGT and a wild-type HCC case (b).

mortality rates for primary liver and intrahepatic bile ductcancer were 4.8/100.000 inhabitants in 2000 and 5.0/100.000inhabitants in 2001 [5, 31]. Moreover, the mortality rates forthese cancers in Antioquia state increased from 6.9/100.000in 2003 to 33/100.000 in 2005 [32]. The increased trend inmortality rates for the country and for Antioquia state couldbe related to improved diagnostic procedures in the healthsystem. On the other hand, changes in risk factors over timecould also be implicated.

The analysis of GLOBOCAN data revealed an overallmale : female ratio of 2.4 : 1. The reported ratios usuallyvaried between 2 : 1 to 4 : 1 depending on the incidence ratesand risk factor patterns over the world. The higher rates ofliver cancer in male population could be due to differences inrisk factor exposure [33].

Central and South America have the lowest reportedmale : female ratio for liver cancer; for example, 1.2 : 1in Colombia and 1.6 : 1 in Costa Rica [33]. However, asmentioned previously the data sources from Colombia in theGLOBOCAN database are restricted to the cancer registry ofCali city. In this study, from the 202 HCC cases, diagnosedat 4 institutions in Bogota, Medellin, and Cali cities, themale : female incidence ratio was 1.8 : 1.

In low-risk populations, the highest age-specific ratesarise in patients around 75 years old while in high-riskpopulation it occurs around 60 to 65 years old. The age-specific pattern is related to differences in HBV and HCV

prevalence, age of infection, and other relevant risk factorsin a population [33]. The mean age of the 202 HCC casesof this study was 62 years and the median 61 years. Thesedata are similar to the median age in other studies carriedout in Peru [34] and Argentina [13] and also in a multicenterprospective HCC study in 9 Latin American countries [6].Nevertheless, in other studies the mean age of HCC patientswas 41.4 years in the Peruvian population [35], 55.9 years inthe Brazilian population [12], and 56 years in the Chileanpopulation [14]. The mean age difference among the HCCstudies in Latin American countries could be related to therisk factor patterns in each country.

According to the IARC TP53 database [36] (R15,http://www-p53.iarc.fr), TP53 mutations have beendescribed in up to 31.4% of HCC cases, with the 249ser

mutation being the most common. This mutation has beenassociated with AFB1 exposition and there is robust evidencethat supports this finding. AFB1 is classified as an IARCGroup 1 carcinogen for the liver and causes an inactivatingmutation at codon 249 of the TP53 tumor suppressor gene,inducing the substitution of an arginine by a serine (R249Smutation) [1].

This mutation has a high frequency in populationsfrom the highest-HCC-incidence areas like Qidong, China(43.8%) [37], Guangxi, China (36%) [38], and Gambia(39.8%) [39]. Moreover, 71.5% of reported mutations incodon 249 of TP53, which correspond to the transversion


(a) (b)

(c) (d)

(e) (f)

Figure 2: Immunohistochemical detection of HCC biomarkers. (a) Liver tissue specimens from Sprague-Dawley rats treated with1.25 mg/Kg of AFB. (b) Liver sample from patient with HCC without apparent exposure to AFB. (c) HCC case with detection of HBxprotein, high expression, and nuclear localization. (d) Negative HCC case to HBx protein detection. (e) HCC case with detection of HCVCore protein, high expression in the cytoplasm of liver cells. (f) HCC case without expression of HCV core protein. Original magnification×400.

G : C → T : A have been detected in HCC [36], wherebythis mutation could be a biomarker for dietary exposure toanatoxin.

In this study, the 249ser mutation was detected in 10.5%(4/38) of the HCC samples. A similar prevalence wasreported in other areas in Anhui, Province of China (10.5%)[40], India (9.5%) [41], and Taiwan (13%) [42]; in recent

studies a prevalence around 2% was reported in Turkey [43]and in Taiwan [44]. The main risk factor of HCC in thesestudies was HBV infection, similar to that reported in thisColombian population.

The 249ser mutation frequency described for the first timein a Colombian population suggests an AFB1 exposure levelbetween low to intermediate. Three of the four HCC samples


F

H

G

A

B

E

C

D

86

114

AY311370 F3

X75663 F.

.

DQ899150 F3.

AB166850 F4.

AF223962 F4.

DQ899142 F2a.

DQ899145 F2b.

AB116552 F1b.

AY090461 F1a.

AF369535 H.

AY090454 H.

AB056513 G.

AB064310 G.

EF397975 A.

X51970 A.

D00330 B.

D00331 B.

AB091256 E.

X75664 E.

EF397978 C.

X75665 C.

36

EF397974 D.

AB033559 D.

0.01

Figure 3: Phylogenetic tree of HBV genotypes generated byNeighbour joining method (MEGA), using HBV S gene sequences.The accesion number followed by genotype indentity is indicated.Bootstrap values are shown (1000 repetitions). HKY was used toaccess distances. Red arrows show the Colombian strain position.

positive for this mutation were cases recruited in one of themost important reference national health center (FundacionSanta fe de Bogota); probably these samples corresponded toHCC cases from rural areas.

A correlation between AFB1 adducts and the 249ser wasdemonstrated in two studies carried out in Taiwan [42, 44].However, none of the HCC cases included in our studywere positive for AFB1-DNA adducts, even though there isevidence of AB1 contamination in corn and rice collectedin supermarkets, retail stores, and stock centers in Colombia[19].

Until now, there are only two studies published regardingthe 249ser mutations in HCC cases in Latin Americancountries. One of them was carried out in Mexico in 16 HCCsamples with a 249ser frequency of 19% [45]. The other onewas carried out in Brazil, where the maximum AFB1 level infood allowed is higher than in North America and Europe;

the 249ser prevalence was 28% (21/74) by PCR-RFLP and16% by direct sequencing [46]. The 249ser prevalence data of16% in Brazil and 10.5% in Colombia using both techniques(RFLP and direct sequence) for mutation analysis suggestthat dietary exposure to AFB1 is an HCC risk factor in theseLatin America countries, although not as important as inAfrica and Asia.

Different studies have suggested that 249ser mutationoccurs almost exclusively in the context of chronic HBVinfection in addition to AFB1 chronic dietary exposure [1].Kirk et al. demonstrated in a prospective study in Gambia,a multiplicative effect on HCC risk resulting from HBVchronic infection and the mutational effect of AFB1 on theTP53 gene (codon 249). Actually, in 216 HCC incident casesand 121 cirrhosis cases, the risk for HCC was associatedwith HBV markers with an odds ratio (OR) of 10.0 (95%,CI: 5.16–19.6), 249ser with an OR of 13.2 (95%, CI: 4.99–35.0), and both markers with an OR of 300 (95%, CI: 48.6–3270) [39]. In this study of a Colombian population, theanalysis of HBV biomarkers and TP53 exon 7 sequenceswas performed in 32 HCC samples, including 3 from 4249ser mutation-positive samples. One of three samples waspositive for both biomarkers (HBV and 249ser). The negativeresults for HBV biomarkers in the other two samples could beexplained by technical limitations of immunohistochemistryand PCR protocols used in this study. However, Kirk etal. have described 15.3% of HCC samples HBV(−)/TP53249ser(+) [39]. Additionally, we report two point mutations,G12457T (exon 5) and G13804A (exon 8), located in the p53DNA-binding domain. These missense mutations modifythe stability and transactivating properties of p53 protein;according to the TP53 database, their frequency in HCCis 2.1 and 0.26%, respectively [36]. In this study, themutations were described in two HCC cases (2.9%, 1/34,each one).

HBV biomarkers were demonstrated in 58.1% of theHCC samples analyzed in this study by HBx proteinimmunohistochemistry detection and HBx PCR. These dataare expected considering the epidemiological pattern of HBVinfection in Colombia, average moderate prevalence but withregions of high prevalence [47, 48]. A higher frequency ofHBV infection than other risk factors has also been describedin HCC patients from Peru (38.9%, 42.2%) [34, 35] and instates of the north eastern and northern regions of Brazil(average 43.1%), including Para (71.4%, 5/7), Bahia (45%,9/20), Minas Gerais (37.8%, 14/37), and Espirito Santo(41.6%, 10/24) [12]. HBV infection was diagnosed in thesestudies by the detection of HBsAg in serum samples fromthe patients. As mentioned previously, data on HBsAg inthe Colombian patients were not available from the clinicalrecords.

The phylogenetic study of the S sequence in three HCCsamples in this study was successful. Two isolates werecharacterized as genotype F and one as genotype D. TheHBV genotypes in the Colombian population have beencharacterized in four studies in blood donor populations,pregnant women, and recently in end-stage liver diseasescases. The predominance of genotype F was demonstratedin three of these studies (77%, 87.23%, and 100%) [49–51].


Additionally, genotypes A, D, C, G, and E were also describedin some cases of those studies [49, 50, 52].

The HCV infection status was established by HCVCore protein immunohistochemistry detection. Regrettably,molecular biology markers for HCV infection were notavailable. A prevalence of 37% was found in the HCC sam-ples analyzed in this study with an HBV/HCV coinfectionprevalence of 19.2%. In other Latin American countrieswhere the HBV infection is the most important HCCrisk factor, the prevalence of the HCV marker is lower(5.3%–16.6%) than described in this Colombian population.Nevertheless, the HCV biomarker in these studies carried outin Peru [34, 35] and Brazil [12] corresponded to anti-HCVdetection; unfortunately, data on this biomarker were notavailable for the Colombian population in this study.

On the other hand, HCV infection is the predominantHCC risk factor in Argentina, Chile, and the South easternstates of Brazil (Rio de Janeiro, Sao Paulo, Parana, andRio Grande do Sul). The prevalence range of the serumanti-HCV marker was 32.8% to 44% in HCC cases fromthese South American countries [12–14]. Furthermore,in a recent prospective multicenter study of HCC casesfrom 9 Latin American countries, including Argentina,Brazil, Chile, Colombia, Uruguay, and Venezuela, the mainHCC risk factor was HCV infection (30.8%), followedby alcohol (20.4%), HBV infection (10.8%), and HCVplus alcohol (5.8%) [6]. Some cases from this multicenterstudy corresponded to patients recruited in a prospectivestudy carried out at Hospital Pablo Tobon Uribe (HPTU)in Medellin city; Interestingly, from the total 131 casesof end-stage liver disease in this hospital in the periodof the study, the most important risk factor was chronicalcoholism (37.4%, 49/131), whereas viral infection rankedsecond (10.7% HBV and 6.9% HCV); the serum markersfor diagnosis of viral infections were HBsAg and anti-HCV [51]. Chronic alcoholism was also an important riskfactor in HCC patients in Argentina (42%), Brazil (37%),and Chile (31%) [12–14]. These results suggest that theHCC risk factors pattern in Latin America is changing tothe pattern seen in developed countries, where one of theprinciple HCC risk factors is chronic alcohol abuse [1].However, further investigations are necessary to confirm thishypothesis.

Unfortunately, only demographic and histopathologicaldata but not alcohol intake were available from the HCCcases in this study. The discrepancy between the risk factorpattern revealed in this retrospective study carried out infour Departments of Pathology from health institutions inCali, Medellin, and Bogota cities and in the prospectiveone performed at hepatology unit of HPTU in Medellincity could be partially explained by differences in agerange, male : female ratio, viral infection biomarkers, andquality of clinical information, including follow-up of thepatients in the second study. However, the predominance ofHBV infection compared to HCV infection was describedin both populations. Further case-control studies in theColombian population are necessary in order to define theburden of alcohol intake and viral infections as HCC riskfactors.

In conclusion, this first retrospective study exploringthe epidemiological features of HCC in patients from 4leading health institutions in the three most important citiesin Colombia revealed that the majority of the patients aremale and in their 6th to 7th decade of life. The main HCCrisk factor is HBV infection but the presence of codon 249mutations, a biomarker of AFB1 exposure, was also found insome HCC samples.

Conflict of Interests

The authors declared no conflict of interests.

Acknowledgments

The authors would like to thank Dr. Anne Lise Haenni forpaper review. This study was supported by DepartamentoNacional de Ciencia Innovacion y Tecnologia Colciencias(Grant 115 041 6445), ECOS Nord (Grant C09S01), Univer-sity of Antioquia, and National Institutes of Health (USA)ES005116 and ES009089.

References

[1] P. Peter and B. Levin, Eds., World Cancer Report 2008,International Agency for Research on Cancer, 2008.

[2] J. F. Perz, G. L. Armstrong, L. A. Farrington, Y. J. Hutin,and B. P. Bell, “The contributions of hepatitis B virus andhepatitis C virus infections to cirrhosis and primary livercancer worldwide,” Journal of Hepatology, vol. 45, no. 4, pp.529–538, 2006.

[3] A. P. Venook, C. Papandreou, J. Furuse, and L. L. deGuevara, “The incidence and epidemiology of hepatocellularcarcinoma: a global and regional perspective,” The Oncologist,vol. 15, supplement 4, pp. 5–13, 2010.

[4] International Agency for Research on Cancer, Cancer Incidencein Five Continents, vol. 9, 2007.

[5] COLOMBIA. DANE (Departamento Administrativo Nacionalde Estadıstica). Direccion de Censos y Demografıa. Bogota,2001.

[6] E. Fassio, S. Dıaz, C. Santa et al., “Etiology of hepatocellularcarcinoma in Latin America: a prospective, multicenter,international study,” Annals of Hepatology, vol. 9, no. 1, pp.63–69, 2010.

[7] M. S. Anez, P. Zabaleta, L. Umbria et al., “Carcinomahepatocellular en los Andes Venezolanos,” GEN, vol. 40, no.2, pp. 66–70, 1986.

[8] R. Mondragon, F. Ochoa, J. Ruiz, R. H. Goepfer, L. F.O.Ocana, and V. A. Crocifoglio, “Carcinoma hepatocelular.Experiencia en el Instituto Nacional de Cancerologıa,” Revistade Gastroenterologia, vol. 62, no. 1, pp. 34–40, 1997.

[9] J. O. Curciarello, L. D. Velez, J. D. Bosia et al., “Carcinomahepatocelular: caracterısticas clinicoepidemiologicas,” ActaGastroenterologica Latinoamericana, vol. 28, no. 3, pp. 243–247, 1998.

[10] H. R. Bueno, S. Indacochea, R. B. Oberst, and G. Chauca, “Rolde los virus de la hepatitis B and Delta en la etiologıa del hepa-tocarcinoma y otras cronicas,” Revista de Gastroenterologia delPeru, vol. 14, no. 2, pp. 135–139, 1994.

[11] A. A. Mattos, S. B. Machado, M. L. Lucas, and C. G. Zettler,“Expresao da proteına mutente do gene supressor tumoral p53


em carcinoma hepatocelular no Sul do brasil: estudo imuno-histoquimico,” Gastroenterologia Endoscopia Digestive, vol. 18,pp. 197–203, 1999.

[12] C. S. Goncalves, F. E. Pereira, and L. C. Gayotto, “Hepa-tocellular carcinoma in Brazil: report of a national survey(Florianopolis, SC, 1995),” Revista do Instituto de MedicinaTropical de Sao Paulo, vol. 39, no. 3, pp. 165–170, 1997.

[13] E. Fassio, C. Mıguez, S. Soria et al., “Etiology of hepatocellularcarcinoma in Argentina: results of a multicenter retrospectivestudy,” Acta Gastroenterologica Latinoamericana, vol. 39, no. 1,pp. 47–52, 2009.

[14] M. Gabrielli, M. Vivanco, J. Hepp et al., “Liver transplantationresults for hepatocellular carcinoma in Chile,” TransplantationProceedings, vol. 42, no. 1, pp. 299–301, 2010.

[15] F. de la Hoz, L. Perez, J. G. Wheeler, M. de Neira, and A. J.Hall, “Vaccine coverage with hepatitis B and other vaccinesin the Colombian Amazon: do health worker knowledgeand perception influence coverage?” Tropical Medicine andInternational Health, vol. 10, no. 4, pp. 322–329, 2005.

[16] F. de la Hoz, L. Perez, M. de Neira, and A. J. Hall, “Eight yearsof hepatitis B vaccination in Colombia with a recombinantvaccine: factors influencing hepatitis B virus infection andeffectiveness,” International Journal of Infectious Diseases, vol.12, no. 2, pp. 183–189, 2008.

[17] M. Beltran, M. C. Navas, F. de la Hoz et al., “Hepatitis C virusseroprevalence in multi-transfused patients in Colombia,”Journal of Clinical Virology, vol. 34, supplement 2, pp. S33–S38, 2005.

[18] S. Duarte-Vogel and L. C. Villamil-Jimenez, “Micotoxinas ensalud publica,” Revista de Salud Publica, vol. 8, supplement 1,pp. 129–135, 2006.

[19] G. J. Diaz, N. S. Perilla, and Y. Rojas, “Ocurrence of aflatoxinsin selected Colombian foods,” Mycotoxin Research, vol. 17, no.1, pp. 15–20, 2001.

[20] I. Gaviria and J. Giraldo, in Evaluacion Genotoxica de Aflatox-inas Aisladas de Maız Almacenado para Consumo Humano yAnimal, Facultad de Ciencias Exactas y Naturales. Universidadde Antioquia, Antioquia, Colombia, 1995.

[21] M. Carpintero, N. Ruiz, and N. Pena, “Niveles de aflatoxina B1en dos zonas del paıs. Primera cosecha de 1979,” Revista ICA,vol. 3, 1980.

[22] O. Cardenas, “Contaminacion de aflatoxinas en productosagrıcolas nacionales. Sorgo del Meta,” Istituto Italiano diTecnologia, vol. 161, no. 26, pp. 41–47, 1986.

[23] G. Prado, “Incidencia de aflatoxina B1 em alimentos,” Revistade Farmacia e Bioquımica, vol. 5, no. 2, pp. 147–157, 1983.

[24] L. L. Hsieh, S. W. Hsu, D. S. Chen, and R. M. Santella,“Immunological detection of aflatoxin B1-DNA adductsformed in vivo,” Cancer Research, vol. 48, no. 22, pp. 6328–6331, 1988.

[25] K. Szymanska, O. A. Lesi, G. D. Kirk et al., “Ser-249 TP53mutation in tumour and plasma DNA of hepatocellularcarcinoma patients from a high incidence area in the Gambia,West Africa,” International Journal of Cancer, vol. 110, no. 3,pp. 374–379, 2004.

[26] G. Hosny, N. Farahat, and P. Hainaut, “TP53 mutationsin circulating free DNA from Egyptian patients with non-Hodgkin’s lymphoma,” Cancer Letters, vol. 275, no. 2, pp. 234–239, 2009.

[27] G. B. Zeng, S. J. Wen, Z. H. Wang, L. Yan, J. Sun, and J. L.Hou, “A novel hepatitis B virus genotyping system by usingrestriction fragment length polymorphism patterns of S geneamplicons,” World Journal of Gastroenterology, vol. 10, no. 21,pp. 3132–3136, 2004.

[28] S. Schaefer, D. Glebe, U. C. Wend, J. Oyunbileg, and W.H. Gerlich, “Universal primers for real-time amplification ofDNA from all known orthohepadnavirus species,” Journal ofClinical Virology, vol. 27, no. 1, pp. 30–37, 2003.

[29] H. Norder, B. Hammas, S. D. Lee et al., “Genetic relatednessof hepatitis B viral strains of diverse geographical origin andnatural variations in the primary structure of the surfaceantigen,” Journal of General Virology, vol. 74, no. 7, pp. 1341–1348, 1993.

[30] N. Mendez-Sanchez, A. R. Villa, G. Vazquez-Elizondo, G.Ponciano-Rodrıguez, and M. Uribe, “Mortality trends for livercancer in Mexico from 2000 to 2006,” Annals of Hepatology,vol. 7, no. 3, pp. 226–229, 2008.

[31] F. L. O. Jaramillo and L. P. M. Velez, “Mortalidad por canceren Colombia 2001,” CES Medicina, vol. 18, no. 2, pp. 19–36,2004.

[32] Direccion Seccional de Salud de Antioquia, “Registro pobla-cional de cancer. Diez primeras causas de mortalidadpor grupos de edad segun 105 grupos,” Departamentode Antioquia 2005, http://www.dssa.gov.co/index.php/esta-disticas/registro-poblacional-de-cancer.

[33] H. B. El-Serag and K. L. Rudolph, “Hepatocellular carcinoma:epidemiology and molecular carcinogenesis,” Gastroenterol-ogy, vol. 132, no. 7, pp. 2557–2576, 2007.

[34] C. B. Sanchez, J. D. Ferrer, R. R. Vargas, M. D. Moscol, andE. Z. Villena, “Clinical—epidemiological characteristics of thehepatocellular carcinoma and treatment in the departamentof digestive system diseases of the National Hospital “EduardoRebagliatti Martins” (HNERM)—ESSALUD,” Revista de Gas-troenterologıa del Peru, vol. 29, no. 1, pp. 17–23, 2009.

[35] E. Ruiz, J. Sanchez, J. Celis et al., “Short and long-termresults of liver resection for hepatocarcinoma in Peru: aPeruvian single center experience on 232 cases,” Revista deGastroenterologia del Peru, vol. 27, no. 3, pp. 223–237, 2007.

[36] A. Petitjean, E. Mathe, S. Kato et al., “Impact of mutant p53functional properties on TP53 mutation patterns and tumorphenotype: lessons from recent developments in the IARCTP53 database,” Human Mutation, vol. 28, no. 6, pp. 622–629,2007.

[37] I. C. Hsu, R. A. Metcalf, T. Sun, J. A. Welsh, N. J. Wang, andC. C. Harris, “Mutational hotspot in the p53 gene in humanhepatocellular carcinomas,” Nature, vol. 350, no. 6317, pp.427–428, 1991.

[38] M. C. Stern, D. M. Umbach, M. C. Yu, S. J. London, Z.Q. Zhang, and J. A. Taylor, “Hepatitis B, aflatoxin B1, andp53 codon 249 mutation in hepatocellular carcinomas fromGuangxi, People’s Republic of China, and a Meta-analysisof existing studies,” Cancer Epidemiology Biomarkers andPrevention, vol. 10, no. 6, pp. 617–625, 2001.

[39] G. D. Kirk, O. A. Lesi, M. Mendy et al., “249ser TP53mutation in plasma DNA, hepatitis B viral infection, and riskof hepatocellular carcinoma,” Oncogene, vol. 24, no. 38, pp.5858–5867, 2005.

[40] H. Liu, Y. Wang, Q. Zhou, S. Y. Gui, and X. Li, “The pointmutation of p53 gene exon7 in hepatocellular carcinoma fromAnhui province, a non HCC prevalent area in China,” WorldJournal of Gastroenterology, vol. 8, no. 3, pp. 480–482, 2002.

[41] S. Katiyar, B. C. Dash, V. Thakur, R. C. Guptan, S. K. Sarin,and B. C. Das, “p53 tumor suppressor gene mutations inhepatocellular carcinoma patients in India,” Cancer, vol. 88,no. 7, pp. 1565–1573, 2000.

[42] R. M. Lunn, Y. J. Zhang, L. Y. Wang et al., “p53 mutations,chronic hepatitis B virus infection, and aflatoxin exposure in


hepatocellular carcinoma in Taiwan,” Cancer Research, vol. 57,no. 16, pp. 3471–3477, 1997.

[43] F. T. Ozdemir, A. Tiftikci, S. Sancak et al., “The prevalenceof the mutation in codon 249 of the P53 gene in patientswith hepatocellular carcinoma (HCC) in Turkey,” Journal ofGastrointestinal Cancer, vol. 41, no. 3, pp. 185–189, 2010.

[44] Y. J. Zhang, P. Rossner Jr, Y. Chen et al., “Aflatoxin B1 andpolycyclic aromatic hydrocarbon adducts, p53 mutations andp16 methylation in liver tissue and plasma of hepatocellularcarcinoma patients,” International Journal of Cancer, vol. 119,no. 5, pp. 985–991, 2006.

[45] Y. Soini, S. C. Chia, W. P. Bennett et al., “An aflatoxin-associated mutational hotspot at codon 249 in the p53 tumorsuppressor gene occurs in hepatocellular carcinomas fromMexico,” Carcinogenesis, vol. 17, no. 5, pp. 1007–1012, 1996.

[46] J. A. Nogueira, S. K. Ono-Nita, M. E. Nita et al., “249 TP53mutation has high prevalence and is correlated with larger andpoorly differentiated HCC in Brazilian patients,” BMC Cancer,vol. 9, no. 204, 2009.

[47] Sivigila, “Sistema de vigilancia en salud publica,” Inf QuincEpidemiol Nac, vol. 11, pp. 49–64, 2006.

[48] Sivigila, “Sistema de vigilancia en salud publica,” Inf QuincEpidemiol Nac, vol. 11, pp. 297–312, 2006.

[49] M. Devesa, C. L. Loureiro, Y. Rivas et al., “Subgenotype diver-sity of hepatitis B Virus American genotype F in amerindiansfrom Venezuela and the general population of Colombia,”Journal of Medical Virology, vol. 80, no. 1, pp. 20–26, 2008.

[50] M. V. A. Mora, C. M. Romano, M. S. Gomes-Gouveaet al., “Molecular characterization of the hepatitis B virusgenotypes in Colombia: a bayesian inference on the genotypeF,” Infection, Genetics and Evolution, vol. 11, no. 1, pp. 103–108,2011.

[51] F. Cortes-Mancera, C. L. Loureiro, S. Hoyos et al., “Etiologyand viral genotype in patients with end-stage liver diseasesadmitted to a Hepatology unit in Colombia,” HepatitisResearch and Treatment, vol. 2011, Article ID 363205, 10 pages,2011.

[52] M. V. A. Mora, C. M. Romano, M. S. Gomes-Gouvea, M. F.Gutierrez, F. J. Carrilho, and J. R. R. Pinho, “Molecular epi-demiology and genetic diversity of hepatitis B virus genotype Ein an isolated Afro-Colombian community,” Journal of GeneralVirology, vol. 91, no. 2, pp. 501–508, 2010.

Submit your manuscripts athttp://www.hindawi.com

Stem CellsInternational

Hindawi Publishing Corporationhttp://www.hindawi.com Volume 2014


MEDIATORSINFLAMMATION

of


Behavioural Neurology

EndocrinologyInternational Journal of



Disease Markers


BioMed Research International

OncologyJournal of



Oxidative Medicine and Cellular Longevity


PPAR Research

The Scientific World JournalHindawi Publishing Corporation http://www.hindawi.com Volume 2014

Immunology ResearchHindawi Publishing Corporationhttp://www.hindawi.com Volume 2014

Journal of

ObesityJournal of



Computational and Mathematical Methods in Medicine

OphthalmologyJournal of


Diabetes ResearchJournal of



Research and TreatmentAIDS


Gastroenterology Research and Practice


Parkinson’s Disease

Evidence-Based Complementary and Alternative Medicine

Volume 2014Hindawi Publishing Corporationhttp://www.hindawi.com

hepatitisbandhepatitiscinfectionbiomarkersand...

Documents